PFAS ghosts: how to identify, evaluate, and exorcise new and existing analytical interference

With increasing public awareness of PFAS, and their presence in biological and environmental media across the globe, comes a matching increase in the number of PFAS monitoring studies. As more matrices and sample cohorts are examined, there are more opportunities for matrix interferents to appear as PFAS where there are none (i.e., "seeing ghosts"), impacting subsequent reports. Addressing these ghosts is vital for the research community, as proper analytical measurements are necessary for decision-makers to understand the presence, levels, and potential risks associated with PFAS and protect human and environmental health. To date, PFAS interference has been identified in several matrices (e.g., food, shellfish, blood, tissue); however, additional unidentified interferents are likely to be observed as PFAS research continues to expand. Therefore, the aim of this commentary is several fold: (1) to create and support a publicly available dataset of all currently known PFAS analytical interferents, (2) to allow for the expansion of that dataset as more sources of interference are identified, and (3) to advise the wider scientific community on how to both identify and eliminate current or new analytical interference in PFAS analyses.

Mixture Effects of Per- and Polyfluoroalkyl Substances on Embryonic and Larval Sheepshead Minnows (Cyprinodon variegatus)

Per- and polyfluoroalkyl substances (PFAS) are ubiquitous and persistent environmental contaminants originating from many everyday products. Perfluorooctane sulfonic acid (PFOS) and perfluorooctanoic acid (PFOA) are two PFAS that are commonly found at high concentrations in aquatic environments. Both chemicals have previously been shown to be toxic to fish, as well as having complex and largely uncharacterized mixture effects. However, limited information is available on marine and estuarine species. In this study, embryonic and larval sheepshead minnows (Cyprinodon variegatus) were exposed to several PFAS mixtures to assess lethal and sublethal effects. PFOS alone was acutely toxic to larvae, with a 96 h LC50 of 1.97 mg/L (1.64-2.16). PFOS + PFOA resulted in a larval LC50 of 3.10 (2.62-3.79) mg/L, suggesting an antagonistic effect. These observations were supported by significant reductions in malondialdehyde (105% ± 3.25) and increases in reduced glutathione concentrations (43.8% ± 1.78) in PFOS + PFOA exposures compared to PFOS-only treatments, indicating reduced oxidative stress. While PFOA reduced PFOS-induced mortality (97.0% ± 3.03), perfluorohexanoic acid (PFHxA) and perfluorobutanoic acid (PFBA) did not. PFOS alone did not affect expression of peroxisome proliferator-activated receptor alpha (pparα) but significantly upregulated apolipoprotein A4 (apoa4) (112.4% ± 17.8), a downstream product of pparα, while none of the other individually tested PFAS affected apoa4 expression. These findings suggest that there are antagonistic interactions between PFOA and PFOS that may reduce mixture toxicity in larval sheepshead minnows through reduced oxidative stress. Elucidating mechanisms of toxicity and interactions between PFAS will aid environmental regulation and management of these ubiquitous pollutants.

Target and Suspect Screening Integrated with Machine Learning to Discover Per- and Polyfluoroalkyl Substance Source Fingerprints

This study elucidates per- and polyfluoroalkyl substance (PFAS) fingerprints for specific PFAS source types. Ninety-two samples were collected from aqueous film-forming foam impacted groundwater (AFFF-GW), landfill leachate, biosolids leachate, municipal wastewater treatment plant effluent (WWTP), and wastewater effluent from the pulp and paper and power generation industries. High-resolution mass spectrometry operated with electrospray ionization in negative mode was used to quantify up to 50 target PFASs and screen and semi-quantify up to 2,266 suspect PFASs in each sample. Machine learning classifiers were used to identify PFASs that were diagnostic of each source type. Four C5-C7 perfluoroalkyl acids and one suspect PFAS (trihydrogen-substituted fluoroethernonanoic acid) were diagnostic of AFFF-GW. Two target PFASs (5:3 and 6:2 fluorotelomer carboxylic acids) and two suspect PFASs (4:2 fluorotelomer-thia-acetic acid and N-methylperfluoropropane sulfonamido acetic acid) were diagnostic of landfill leachate. Biosolids leachates were best classified along with landfill leachates and N-methyl and N-ethyl perfluorooctane sulfonamido acetic acid assisted in that classification. WWTP, pulp and paper, and power generation samples contained few target PFASs, but fipronil (a fluorinated insecticide) was diagnostic of WWTP samples. Our results provide PFAS fingerprints for known sources and identify target and suspect PFASs that can be used for source allocation.

Disposition of Fluorine on New Firefighter Turnout Gear

Firefighter turnout gear is essential for reducing occupational exposure to hazardous chemicals during training and fire events. Per-and polyfluoroalkyl substances (PFASs) are observed in firefighter serum, and possible occupational sources include the air and dust of fires, aqueous film-forming foam, and turnout gear. Limited data exist for nonvolatile and volatile PFASs on firefighter turnout gear and the disposition of fluorine on the individual layers of turnout gear. Further implications for exposure to fluorine on turnout gear are not well understood. Three unused turnout garments purchased in 2019 and one purchased in 2008, were analyzed for 50 nonvolatile and 15 volatile PFASs by liquid chromatography quadrupole time-of-flight mass spectrometry (LC-qTOF-MS) and gas chromatography-mass spectrometry (GC-MS), respectively. Particle-induced gamma ray emission (PIGE), a surface technique, and instrumental neutron activation analysis (INAA), a bulk technique, were used to measure total fluorine. Bulk characterization of the layers by pyrolysis-GC/MS (py-GC/MS) was used to differentiate fluoropolymer (e.g., PTFE) films from textile layers finished with side-chain polymers. The outer layer, moisture barrier, and thermal layers of the turnout gear all yielded measured concentrations of volatile PFASs that exceeded nonvolatile PFAS concentrations, but the summed molar concentrations made up only a small fraction of total fluorine (0.0016-6.7%). Moisture barrier layers comprised a PTFE film, as determined by py-GC-MS, and gave the highest individual nonvolatile (0.159 mg F/kg) and volatile PFAS (20.7 mg F/kg) as well as total fluorine (122,000 mg F/kg) concentrations. Outer and thermal layers comprised aromatic polyamide-based fibers (aramid) treated with side-chain fluoropolymers and had lower levels of individual nonvolatile and volatile PFASs. Equal concentrations of total fluorine by both PIGE and INAA on the outer and thermal layers is consistent with treatment with a side-chain fluoropolymer coating. New turnout gear should be examined as a potential source of firefighter occupational exposure to nonvolatile and volatile PFASs in future assessments.

Environmental Source Tracking of Per- and Polyfluoroalkyl Substances within a Forensic Context: Current and Future Techniques

The source tracking of per- and polyfluoroalkyl substances (PFASs) is a new and increasingly necessary subfield within environmental forensics. We define PFAS source tracking as the accurate characterization and differentiation of multiple sources contributing to PFAS contamination in the environment. PFAS source tracking should employ analytical measurements, multivariate analyses, and an understanding of PFAS fate and transport within the framework of a conceptual site model. Converging lines of evidence used to differentiate PFAS sources include: identification of PFASs strongly associated with unique sources; the ratios of PFAS homologues, classes, and isomers at a contaminated site; and a site's hydrogeochemical conditions. As the field of PFAS source tracking progresses, the development of new PFAS analytical standards and the wider availability of high-resolution mass spectral data will enhance currently available analytical capabilities. In addition, multivariate computational tools, including unsupervised (i.e., exploratory) and supervised (i.e., predictive) machine learning techniques, may lead to novel insights that define a targeted list of PFASs that will be useful for environmental PFAS source tracking. In this Perspective, we identify the current tools available and principal developments necessary to enable greater confidence in environmental source tracking to identify and apportion PFAS sources.

Utilization of a NIST SRM: a case study for per- and polyfluoroalkyl substances in NIST SRM 1957 organic contaminants in non-fortified human serum

The National Institute of Standards and Technology (NIST) generates and maintains thousands of Standard Reference Materials (SRMs) to serve commerce worldwide. Many SRMs contain metrologically traceable mass fractions of known organic chemicals and are commercially available to aid the analytical chemistry community. One such material, NIST SRM 1957 Organic Contaminants in Non-Fortified Human Serum, was one of the first materials issued by NIST with measurements for per- and polyfluoroalkyl substances (PFAS) listed on the Certificate of Analysis and was commercially available in 2009. Since the release of SRM 1957, nearly 400 units have been sold to date, and over 50 publications related to PFAS measurements have included this material for multiple analytical purposes, such as a quality control material, for interlaboratory comparison, as an in-house comparison tool, for inter- and intra-day measurement accuracy, as an indicator of isomeric patterns of PFAS, and for other uses. This perspective details the ways SRM 1957 is utilized by the analytical community and how data have been reported in the literature. A discussion on accurately comparing SRM data to generated data is included. Furthermore, we conducted an in-depth investigation around additional applications for NIST SRMs, such as a matrix-matched reference material, and for the identification of targeted compounds during high-resolution mass spectrometry data collection. Ultimately, this manuscript illustratively describes the ways to utilize a NIST SRMs for chemicals of emerging concern.

Spatial Trends of Anionic, Zwitterionic, and Cationic PFASs at an AFFF-Impacted Site

Soil and groundwater from an aqueous film-forming foam (AFFF)-impacted site were sampled at high resolution (n = 105 for soil, n = 58 for groundwater) and analyzed for an extensive list of anionic, zwitterionic, and cationic poly- and perfluoroalkyl substances (PFASs). Spatial trends for perfluoroalkyl acids and many precursors enabled a better understanding of PFAS composition, transport, and transformation. All PFASs without analytical standards were semi-quantified. Summed PFAS and individual PFAS concentrations were often higher at depth than near the surface in soil and groundwater. Zwitterionic and cationic compounds composed a majority of the total PFAS mass (up to 97%) in firefighter training area (FTA) soil. Composition of PFAS class, chain length, and structural isomers changed with depth and distance from the FTA, suggesting in situ transformation and differential transport. The percentage of branched perfluorooctane sulfonate increased with depth, consistent with differential isomeric transport. However, linear perfluorooctanoic acid (PFOA) was enriched, suggesting fluorotelomer precursor transformation to linear PFOA. Perfluorohexane sulfonamide, a potential transformation product of sulfonamide-based PFASs, was present at high concentrations (maximum 448 ng/g in soil, 3.4 mg/L in groundwater). Precursor compounds may create long-term sources of perfluoroalkyl acids, although many pathways remain unknown; precursor analysis is critical for PFAS fate and transport understanding.